# Chengcheng P.

> model inference and fine tuning

Location: Santa Clara, California, United States
Profile: https://flows.cv/chengcheng

## Work Experience
### Staff Software Engineer @ Voyage AI by MongoDB
Jan 2025 – Present
Working on llm inference efficiency: weights loading, batching, inference profiling/optimization etc

1, https://www.mongodb.com/company/blog/engineering/token-count-based-batching-faster-cheaper-embedding-inference-for-queries

2, Led and coordinated eng efforts serving voyage 4 series models: https://huggingface.co/voyageai

3, Profiled and tuned MFU of Our MoE Model

4, Onboarded voyage-4-nano to vllm

5, [WIP] one gpu switch models in 1 second

### Software Engineer @ ZettaBlock
Jan 2023 – Jan 2025
AI and ML infrastructure
Big data infrastructure

### Senior Software Engineer @ Uber
Jan 2021 – Jan 2023
Data Infra and platform
ML infra

### Software Engineer, Data Infra @ Postmates by Uber
Jan 2019 – Jan 2021
Worked on and oncall all the cool services and infra below:
1, Kafka/ZK cluster and kafka tools/libs
2, Schema registry and serde lib in go/python
3, Seldon-core model serving platform
4, Experiments service
5, Feast feature store

### Software engineer MTS1 @ eBay
Jan 2019 – Jan 2019

### Senior Software Engineer @ eBay
Jan 2018 – Jan 2019
Wrote historical/real-time pipelines in Spark/Hadoop/Hive/ElasticSearch on OpenStack cloud and K8S
Oncall and improved eBay inhouse columnar DB Portico.

### Senior Software Developer @ OpenText
Jan 2017 – Jan 2018

### Application Developer @ eSentire
Jan 2014 – Jan 2017
eSentire acquired by Warburg Pincus

### Teaching Assistant @ University of Waterloo
Jan 2012 – Jan 2014 | Wateroo, ON
Teaching assistant for undergraduate security course, graduate & undergraduate database courses


## Education
### Master's degree
University of Waterloo

### Southeast University


## Contact & Social
- LinkedIn: https://linkedin.com/in/chengcheng-p-53560238

---
Source: https://flows.cv/chengcheng
JSON Resume: https://flows.cv/chengcheng/resume.json
Last updated: 2026-04-01